PartialDependence#

Link to Algorithm description: Partial Dependence Plot

class interpret.blackbox.PartialDependence(model, data, feature_names=None, feature_types=None, num_points=10, std_coef=1.0)#

Partial dependence plots as defined in Friedman’s paper on “Greedy function approximation: a gradient boosting machine”.

Friedman, Jerome H. “Greedy function approximation: a gradient boosting machine.” Annals of statistics (2001): 1189-1232.

Initializes class.

Parameters:
  • model – model or prediction function of model (predict_proba for classification or predict for regression)

  • data – Data used to initialize PartialDependence with.

  • feature_names – List of feature names.

  • feature_types – List of feature types.

  • num_points – Number of grid points for the x axis.

  • std_coef – Co-efficient for standard deviation.

explain_global(name=None)#

Provides approximate global explanation for blackbox model.

Parameters:

name – User-defined explanation name.

Returns:

An explanation object, visualizes dependence plots.